Population size is not the whole story. Composition also matters. Because immigrants have characteristics different from those of native-born Americans, they are changing the nation’s demographic profile. At the end of the day the racial and ethnic shifts triggered by immigration may be more important to the U.S. economy than the population increase itself.
For most of our history whites were a dominant and growing share of our nation’s population. The first census, in 1790, found that almost 81 percent of the country considered itself white. By 1950 that number had risen to 89 percent.1
While whites are still the largest racial group in the country, their grip on that status is at historically low levels and declining rapidly. In 1970, at the beginning of the current wave of immigrants from Asia and Latin America, 83 percent of Americans were white non-Hispanics — descendents of the European immigrants who had arrived over the past 250 years. African Americans — a group descended from African, mostly involuntary immigrants from a previous century — made up 11 percent. Hispanics and Asians made up less than 6 percent.
By 2005 the white non-Hispanic share had shrunk to 67 percent while the black share of U.S. population inched up to 13 percent. The Hispanic and Asian populations soared, reaching a combined 19 percent of U.S. population.
It’s not that U.S. whites declined in numbers over this period, but that minority groups increased more rapidly. Thus from 1990 to 2005 the number of white, non-Hispanics living in the U.S. grew by 9.5 million, or by about 5 percent, while all other races grew by 35.3 million, or 58 percent. Since 1990 non-whites have accounted for nearly 80 percent of U.S. population growth, according to Census data.
[Note: From a global perspective, the U.S. is still an oasis of “whiteness.” Whites are expected to account for only 9 percent of global population in 2010 compared with 17 percent in 1997, according to demographer Harold Hodgkinson. If this projection holds true, whites are now the world’s smallest minority, albeit the largest racial group in the United States.2]
The rapid rise in U.S. Hispanics and Asians is due mainly to immigration. Mexicans accounted for 30.7 percent of all foreign born in 2005, up from 28 percent in 2000, 22 percent in 1990, and 16 percent in 1980. In 2005, Mexico accounted for almost six times as many immigrants as the combined total for China, Taiwan, and Hong Kong. As recently as 1970 Mexicans comprised only the 4th highest foreign-born population — behind Italians, Germans, and Canadians.
If immigration remains at current levels, the racial and ethnic shift will be even greater over the next several decades. This is from the Census Bureau’s analysis of future demographic trends:
The highest levels of net international migration correspond to the largest amount of growth for the Asian and Hispanic populations, which are the primary immigrant groups to the United States. For both the Low and High Net International Migration series, these populations are projected to more than double in size between 2000 and 2050. Even if net international migration is maintained at a constant level of nearly one million, the Hispanic population is still projected to more than double between 2000 and 2050, while the size of the Asian population is projected to increase by 79 percent….3
Non-Hispanic whites are conspicuously missing from the population growth scenario:
…Most race groups are projected to experience a moderate increase in size over the next four decades for all projection series. One exception to this is the non-Hispanic White alone population, which is projected to experience decline in both the Zero and Constant Net International Migration series. Under the assumption of zero net international migration, for example, this population will decrease by about 6 percent from 2000 to 2050.
Under current immigration policy, Hispanics, blacks, and Asians will constitute more than half of U.S. population by mid-century:
By 2050, Hispanic and non-Hispanic minorities will be in the majority. Their combined populations will constitute 53.7 percent of U.S.population compared to just 35.3 percent in 2010.
Meanwhile, the white share of U.S.population will slip from a 64.7 percent majority in 2010 to a 46.3 percent minority in 2050. And because white fertility — the average number of children born to white women over their child-bearing years — is below the level needed to replace deceased whites, the white population will start to decline around 2030 according to Census Bureau projections.
Thus, unlike earlier periods, a falling white population share will also signal a falling white population.
Even if no new immigrants are allowed in, the minority share of the U.S. population will rise because immigrationhas already pushed the minority populationtorecord levels and because non-white mothers (immigrantandnative-bornalike) will give birth to significantly more children than their white counterparts.
Amoratorium, however, would enable white non-Hispanics to retain their majority status past mid-century:
White non-Hispanics will account for 57.8 percent of U.S. population in 2050 under a moratorium versus 46.3 percent under current immigration policy.
Blacks, Asians, and other non-Hispanics will see their mid-century population share rise to 23.5 percent under current immigration policy versus 21.2 percent under a moratorium.
Similarly, a moratorium would slow, but not halt, the Hispanicization of the U.S. population:
The Hispanic share of the U.S.population would rise from 14.3 percent in 2010 to 30.2 percent in 2050 (under current immigration policy) and to 21 percent (with no further immigration).
Because of higher birthrates among Hispanics already here, a 40 year moratorium would not prevent minorities from constituting a majority of the population under age 5 by 2050. If immigration continues, black, Hispanic, and Asian children will become a majority of young children as early as 2019, according to the latest projections.4
Poverty rates: Moratorium vs. current policy
The nation’s poverty rate climbed to 13.2 percent in 2008, up from 12.5 percent in 2007, according to the Census Bureau’s latest annualreport.5 Poverty spiked to its highest level since 1997, portending even larger increases in 2009 and 2010.
Both The Washington Postand The New York Times put the poverty story on the front page. As usual, neither daily mentioned the word “immigration” when listing factors behind the troubling trend. Yet even in the worst downturn since the 1930s, immigration is exacerbating America’s poverty problem.
First, directly: Many immigrants are poor themselves, adding to thepoverty population. Second, indirectly: U.S.-born children of immigrants are often caught in the same economic treadmill as their parents. The persistence of poverty in second- and third-generation immigrants is especially relevant in assessing the impact of a long-term moratorium.
How large a direct role does immigration play? Here are the Census Bureau’s figures for 2007 and 2008:
First the good news (sort of). For the first time in many years, the growth rate of foreign-born poor lagged that of U.S.-born poor. As a result, the immigrant share of America’s poverty population declined slightly — from 16.5 percent in 2007 to 16.4 percent in 2008.
The good comes with caveats. It ignores the American-born minor children of poor immigrant mothers — nearly 3.0 million by some estimates — counted as“U.S. natives”by the Census. This is huge. The number of Hispanics living in poverty grew by 1.1 million last year, or nearly 8 times the 142,000 blacks added to the rolls. In recent years, most Hispanic population growth has been via U.S.-born children rather than immigration.
The bad news: The share of immigrants living in poverty reached a record 17.8 percent in 2008, up from 16.5 percent in 2007. The 2006-08 period saw a significant widening of the poverty gap between natives and immigrants:
Notwithstanding the yawning gap between foreign-born and native-born poverty, it is clear that variations among immigrant groups are greater still. Thus the poverty rate for first generation Hispanic immigrants in a recent year was 24.5 percent, or about 2.5 times that of first generation white, non-Hispanic immigrants (9.8 percent).6
Poverty remains high even in second- and third-generation immigrant households. The poverty rate for non-immigrant Hispanic households, for example, is 18.9 percent, or more than twice the rate for white non-Hispanic non-immigrants.7
How will the demographic shifts brought about by a moratorium change future poverty rates? In answering this, we must know the current poverty rates for persons (immigrant and U.S.-born) in each race and ethnic group. This is from the 2008 Census report:
Hispanics and blacks are both nearly three times more likely to be poor than white non-Hispanics. Even Asians, a group widely deemed to be intellectually gifted and entrepreneurial, are 40 percent more likely to live in poverty than non-Hispanic whites.
It’s clear that a moratorium, by reducing Hispanic, black, and Asian population shares, will lower the U.S. poverty rate. To quantify the impact, we simply multiply the latest (2008) poverty rate for each group by their population shares as projected under the moratorium and current policy scenarios:
Under current immigration policy, the U.S. poverty rate will be 16.3 percent in 2050, up from 13.2 percent in 2008. A moratorium will lower the 2050 poverty to 14.7 percent, about 10 percent below the rate that would have been reached.
These are projections, not forecasts. They assume that poverty rates for each race and ethnic group remain as they were in 2008 (the latest year of data), and that their population shares change as projected in the Census Bureau’s current policy and moratorium scenarios.
We have ignored a potentially large, albeit indirect, fallout from immigration: the impoverishment of native-born non-Hispanic whites and blacks who lose jobs to immigrants. Other things equal, as immigrant workers displace U.S.-born whites and blacks, the poverty rates for these two groups will rise — as will the potential benefits of a moratorium.
Bottom line: We may have understated the anti-poverty benefits of a moratorium. Our projections are probably too conservative.
Moratorium vs. current policy
The 1996 Welfare Reform Act prohibited immigrants who entered the U.S. after August 22, 1996, from receiving most types of public assistance. (The ban is lifted when the immigrant becomes an American citizen.) Nevertheless, most states filled the gap, enacting programs that grant welfare eligibility to new arrivals. As a result, immigrants continue to receive every major welfare program at higher rates than native-born Americans.
Welfare eligibility is closely linked to poverty. We show above that: 1 immigrants are more likely than natives to have income below the poverty line; 2 the variation in poverty rates among immigrant groups is greater than the variation between immigrants and natives; and 3 poverty remains high even in second- and third-generation immigrant households.
To project the long-term impact of immigration policy on welfare usage, it is, therefore, necessary to know dependency rates for all persons (immigrant and U.S.-born) in each race and ethnic group. Census data allows us to do this.
Hispanics are nearly two and one-half times more likely to receive welfare than whites; blacks are more than three times as likely. Although the minority/white recipiency gap narrowed in the years immediately following the 1996 welfare reform, it has expanded since 2003.
Its clear that a moratorium, by reducing Hispanic and black population shares, will lower the national rate of welfare recipiency. To estimate the impact of different immigration policies on future welfare rates, we simply multiply the projected population shares for whites, blacks, and Hispanics under the moratorium and current policy scenarios by the latest (2005) recipiency rate for each group. The graphic on page 66 says it all.
Under current immigration policy, nearly one in five (19.7 percent) U.S. residents will be on welfare in 2050. That represents a 58 percent increase from 2000, when one in eight (12.5 percent) of residents were on welfare.
Under a moratorium, national welfare recipiency will be 17.9 percent in 2050. While that is above the current rate, it is nearly 10 percent below the rate (19.7 percent) that would have been reached under current immigration policy.
Don’t forget a moratorium will reduce U.S. population below levels that would have been reached under current immigration policy. As a result, the number of persons receiving welfare will fall by more than the drop in the recipiency rate would suggest. Here are our projections:
Other things being equal, a 40 year moratorium will reduce the number of persons receiving welfare in 2050 by 28.6 million, or by about one-third (-33.1 percent). Instead of 86.3 million there will be 57.7 million people receiving cash welfare (TANF), Food Stamps, or SSI that year.
The cost savings would be enormous. Consider this: The federal government will spend about $650 billion this year (FY2010) on cash and non-cash benefits to poor and economically distressed persons. This includes Medicaid, housing subsidies, the Earned Income Tax Credit (EITC), and unemployment benefits, as well as the three programs —TANF, food stamps, and SSI — whose recipiency rates we use in the above calculation.
A 33.1 percent reduction in all welfare programs translates to an annual savings of $215 billion — $693 for every man, woman, and child in the country.
This is undoubtedly a conservative estimate. Studies show that the recipiency gap between immigrants and natives is larger for EITC, Medicaid, housing subsidies, and other non-cash programs than it is for the three programs whose recipiency rates we use in these calculations.
Immigrant households have more spells of welfare, and these spells are longer. As a result, the typical immigrant household has a much higher propensity of being ‘permanently’ on welfare….8
Conclusion: A moratorium will reduce both the number of poor and the average welfare expenditure per poor person. ■
1. Sean Scully, “Minorities Gain Ground,” Insight on the News, May 28, 2001.
2. Nat Irwin II, “America’s Increasing Diversity,” Futurist, March/April 2004, page 21.
3. Jennifer M. Ortman and
Christine E. Guarneri,
United States Population Projections: 2000-2050,
Census Bureau analysis, 2009.
4. Sam Roberts, “Projections
put whites in the minority in U.S. by 2050,”
New York Times, December
5. Census Bureau,
Poverty, and Health Insurance Coverage in the United States: 2008,
September 2009. Table 4.
6. Robert Rector, Importing Poverty: Immigration and Poverty in the United States, Heritage Special Report, September 26, 2006. Chart 6.
7. Rector, ibid.
8. George J. and Lynette
Hilton, “Immigration and the Welfare State: Immigrant Participation in
Means-Tested Entitlement Programs,”
Quarterly Journal of Economics, May
1996. pp. 575-604.